This near space weather balloon thing is getting exciting.
The photo above is of the view of Canberra from the proposed 27km altitude that the balloon will reach. Lake Burley Griffin looks like a puddle. I can’t wait to see this from a camera I flew into space myself.
Today I was talking to my electronics teacher about the project and he said that he wanted me to do all of the electronics and leave the balloon bit to him. So that’s exactly what I’m doing. The plan with the electronics is to do the first couple of flights with an android smart phone for the brains and tracking and then gradually move away from that to a more complicated system (using home brewed avionics). While I’m building the first version of the balloon with a smartphone and a spy camera I’ll work on some more elaborate systems using proper chips and electronics. What I hope to have by a couple of weeks or months is a fully functioning tracking and USB hosting development board specially made for the purpose. I’ve already started the design with an atmega328 as the core. I might also design a PCB.
But while I design a better system I still have to get the first version up in the air with the android phone system. The main problem with this project is battery life. From what I’ve read, balloon flights can last anywhere from 3-5 hours. That’s a long time for my smart phone to be turned on and a very long time for my little spy camera. The little spy cameras only carry a small battery with around 300mah of capacity. That translates into around about an hour of recording from what I’ve heard. My phone isn’t much better. It will last about a day if I use it a lot to read other blogs and play games at full screen brightness every now and again. That’s probably about 3 hours of solid use. However, it will last for many days if the screen brightness is turned down and it is not using too many wireless functions.
I remember reading an article in the newspaper (can’t remember which one) a while ago about conserving smartphone life. It said that some of the most power-hungry functions on a phone are GPS and 3G internet. Yeah well, I’m using those constantly and heavily. Looks like I’ll need an external battery source for the phone and the spy camera.
I could go all hacker and solder another battery to the existing connections inside my phone and spy camera but I don’t really feel this is the way to go. The first reason is that it is incredibly easy to fry a phone or spycam by doing this. I tend to fry things because I just plug them in without thinking too hard. Remember my Leanlight project? I blew up two 7805 voltage regulators in that because I wired up the battery the wrong way round. Then I decided it might be time to stop desoldering dead regulators and solder in a protection diode. The second reason for not soldering a battery directly to the device’s existing battery points is that it just looks messy. It’s ugly having an RC plane battery hanging out of a phone or spycam with its case half on. Not neat engineering design. Besides, building a simple dual output USB supply is useful for powering a lot of things that don’t have good battery life on the go (like GoPros etc). Also, a USB power supply can power a dedicated avionics dev board easily with the 5V it will require (I’m going with the good old atmega328 for the dev board and it takes 5V).
For the power supply I decided to simply build two of the Leanlight power supply units on the same board with some modifications. One of the most important things was that I had two of them. Why did I want two when I could just wire both devices up to the same power output? Backup. No-one can hear your 7805 voltage regulator blow up in space. Another important addition was a protection diode. The last thing I need is a 7805 blowing up just when I connect the batteries the wrong way round minutes before launch.
Here’s a pic of what the power supply circuit is. The actual power supply circuit will be a lot more tightly packed. It’s just spread out so you can see the layout.
It is important to remember that this version of the balloon is designed to be simple. I might add some LEDs with some test switches so I can easily see without a multimeter whether the balloon’s electronics power supply is working.
So I’m going to build that power supply board on Friday afternoon or Saturday. Then my electronics work is finished and I can start designing a dedicated dev board. I’ll have to test it in the cold conditions of near space at -50 degrees celsius. I’m going to need a freezer. Maybe the school’s kitchen’s one. Or maybe just mine. I wonder how cold mine gets. Hmmmmm.
Until next time, check out these vids: